get_futures() method works again after removing some class variables

[maneum1]

Looks like following variables

• streamer_symbol,
• is_tradeable,
• future_product

are not anymore provided by a
GET /instruments/futures
request from tastytrade

This is the error I get when using the current released db33a07

# results = symbol_search(s, 'ES')
fut = Future.get_futures(session=s)
print(fut[0])

---------------------------------------------------------------------------
ValidationError                           Traceback (most recent call last)
Cell In[4], line 2
      1 # results = symbol_search(s, 'ES')
----> 2 fut = Future.get_futures(session=s)
      3 print(fut[0])

File ~\python\venvs\env311\Lib\site-packages\tastytrade\instruments.py:666, in Future.get_futures(cls, session, symbols, product_codes)
    662 validate_response(response)
    664 data = response.json()['data']['items']
--> 666 return [cls(**entry) for entry in data]

File ~\python\venvs\env311\Lib\site-packages\tastytrade\instruments.py:666, in <listcomp>(.0)
    662 validate_response(response)
    664 data = response.json()['data']['items']
--> 666 return [cls(**entry) for entry in data]

File ~\python\venvs\env311\Lib\site-packages\pydantic\main.py:150, in BaseModel.__init__(__pydantic_self__, **data)
    148 # `__tracebackhide__` tells pytest and some other tools to omit this function from tracebacks
    149 __tracebackhide__ = True
--> 150 __pydantic_self__.__pydantic_validator__.validate_python(data, self_instance=__pydantic_self__)

ValidationError: 3 validation errors for Future
streamer-symbol
  Field required [type=missing, input_value={'symbol': '/6JU2', 'prod...2', 'symbol': '/6JZ2'}]}, input_type=dict]
    For further information visit https://errors.pydantic.dev/2.1.2/v/missing
is-tradeable
  Field required [type=missing, input_value={'symbol': '/6JU2', 'prod...2', 'symbol': '/6JZ2'}]}, input_type=dict]
    For further information visit https://errors.pydantic.dev/2.1.2/v/missing
future-product
  Field required [type=missing, input_value={'symbol': '/6JU2', 'prod...2', 'symbol': '/6JZ2'}]}, input_type=dict]
    For further information visit https://errors.pydantic.dev/2.1.2/v/missing

after commenting out these 3 Future parameters the get_futures() function works again

...
class Future(TradeableTastytradeJsonDataclass):
    """
    Dataclass that represents a Tastytrade future object. Contains information about
    the future and methods to fetch futures for symbol(s).
    """
...
    # streamer_symbol: str
    back_month_first_calendar_symbol: bool
    # is_tradeable: bool
    # future_product: 'FutureProduct'
...

[Graeme22]

Hey! Thanks for the contribution. So it's not that streamer_symbol or the other keys is never present; rather, they may or may not be present. So the correct fix here would be to make them Optional and give a default value of None, not comment out the fields. Would you be willing to do that?

[maneum1]

Hi Graeme, good thought. they are documented on the Tastytrade API documentation, even though these vars are not returned when fetching Futures. I uncommented them and assigned 'None' to them to make them optional.

This is the first time I made a pull request, is there a way to link the update to this one or do I need to start a new pull request from my latest commit?

Huhh, interesting I just see that my latest commit is already shown above, Cool

[Graeme22]

Haha yup, you just commit to the same branch.

So this will fail the typing tests because None is not a bool nor a string. It needs to be Optional[bool] = None like contract_size. Also, all Optional values should be moved to the bottom of the class, after all non-Optional values.

[maneum1]

hups hehe, Still learning here, ... thanks for the advise to correctly implement this. I hope I got it now right. Please check again

I thought when a variable has a default value assigned in python that this makes it optional. Obviously not entirely correct. ;-)

[Graeme22]

I thought when a variable has a default value assigned in python that this makes it optional. Obviously not entirely correct. ;-)

This looks good! Actually Python doesn't have strong typing, so you're sorta correct, however the pydantic library which this project uses enforces the types.